mirror of
https://github.com/tailscale/tailscale.git
synced 2024-11-27 12:05:40 +00:00
4bbac72868
There are no mutations to the input, so we can support both ~string and ~[]byte just fine. Updates #cleanup Signed-off-by: Joe Tsai <joetsai@digital-static.net>
32 lines
1.1 KiB
Go
32 lines
1.1 KiB
Go
// Copyright (c) Tailscale Inc & AUTHORS
|
|
// SPDX-License-Identifier: BSD-3-Clause
|
|
|
|
// Package truncate provides a utility function for safely truncating UTF-8
|
|
// strings to a fixed length, respecting multi-byte codepoints.
|
|
package truncate
|
|
|
|
// String returns a prefix of a UTF-8 string s, having length no greater than n
|
|
// bytes. If s exceeds this length, it is truncated at a point ≤ n so that the
|
|
// result does not end in a partial UTF-8 encoding. If s is less than or equal
|
|
// to this length, it is returned unmodified.
|
|
func String[String ~string | ~[]byte](s String, n int) String {
|
|
if n >= len(s) {
|
|
return s
|
|
}
|
|
|
|
// Back up until we find the beginning of a UTF-8 encoding.
|
|
for n > 0 && s[n-1]&0xc0 == 0x80 { // 0x10... is a continuation byte
|
|
n--
|
|
}
|
|
|
|
// If we're at the beginning of a multi-byte encoding, back up one more to
|
|
// skip it. It's possible the value was already complete, but it's simpler
|
|
// if we only have to check in one direction.
|
|
//
|
|
// Otherwise, we have a single-byte code (0x00... or 0x01...).
|
|
if n > 0 && s[n-1]&0xc0 == 0xc0 { // 0x11... starts a multibyte encoding
|
|
n--
|
|
}
|
|
return s[:n]
|
|
}
|